In 1975 my father purchased a new International Harvester pickup truck. He used it infrequently to go fishing and rock hunting. He lived in a climate that saw both snow and desert heat. The truck performed flawlessly for many years. When he passed away my daughters used it when they went to the university. It was a big hit with the young men.
After they graduated, it came back to me. This was in 2002, some 27 years later. Over its long life, it only had 67,000 miles on it. I used it in Arizona as a second work truck. Soon after I got it back, it started to just stop at random times. There was neither rhyme nor reason as to when or why.
The 1975 IH pickup – it's in great shape but for a tiny manufacturing flaw.
Every time I started trouble shooting, it worked with perfection. After repeated failures, I began to suspect the ignition system. I replaced all of the ignition components except for the ignition computer module, which was potted with epoxy and sand and was quite expensive. Not that I was likely to find one. The last year the truck was made was around 1977. I finally bit the bullet and replaced the ignition computer. Now I had to wait through several weeks of driving to see it I’d get stranded or if I had really fixed the problem.
I didn’t want to take the risk, so I decided to very carefully remove the sand-filled epoxy to troubleshoot the old computer module. After many hours of picking out small pieces of the mix, I finally discovered the cause of the failure. I couldn’t believe what I found. The computer module had one input -- the 12 VDC from the battery. It had two outputs, one to the ignition coil, and the other was a wire to the metal case of the computer for the ground connection.
The wire to the case was a solid connection, but the connection to the printed circuit board was fed through a solder hole. During manufacture it was never soldered. There was no evidence of any solder at all. The connection was barely touching the edge of the pad. It had been held in place for many years by the epoxy -- an amazing feat. The computer module is now a spare that will most likely never be needed. It goes to show that nothing can be overlooked or assumed.
This entry was submitted by Lee R. Watkins and edited by Rob Spiegel.
Lee R. Watkins has worked for the Westinghouse Surface Division, Honeywell Aerospace, Martin Maritta, Motorola, and ON Semi. He is currently semi-retired and doing consulting. He is a registered P.E. in Colorado and Arizona. He has published several papers on analog filters and a book, Comprehensive Filter Design.
Very cool looking truck. One thing strikes me though--I'm surprised there was any kind of computer module in a truck that looks like an antique. I know it was the 70s and computers did exist, but I'm shocked that there was any kind of electronics on board at that time to even trouble shoot.
My brother had one of these with a Slant 4 (half of a V8) engine in it. He used to call it his "Hickup truck". He drove it in the mountains of Colorado and used to have to get out a lot to adjust the mixture in the carburetor as he went up and down in altitude.
I didn't spend much time looking at it under the hood, but I seem to remember that it was a stone simple and stupid system. My best guess with the "computer" module is that it probably had a few operational amplifiers for adjusting the timing. That would have been a typical thing to call a "computer" in those days.
It was definintely NOT a computer! Just a simple CD ignition system, very popular in that time frame. The idea was to dump a capacitor charged up to a couple hundred volts into the ignition coil; the greatly-reduced rise time and increased peak voltage made it more likely to properly fire even a moderately fouled plug, increasing both performance and fuel mileage. Delta Electronics sold a very popular aftermarket unit you could buy as a kit or fully assembled and tested. I bought the kit, and built it for use in my 1968 Rover 2000TC sports sedan, because the factory Lucas electricals were notorious for unreliability, especially in wet weather. It worked great for quite a while; then the law of unintended consequences bit me! The Rover used carbon ignition wiring for EMI reduction, and I still had the Lucas distributor cap with its tendency to collect condensation inside. Because of the former, and the increase in voltage and faster rise time, I started getting misfiring that I finally traced to a breakdown in the spark cable insulation due to arcing to places where the cables touched grounded metal (valve cover in this case, leading to degradation of the carbon conductors inside the wires. I decided to get rid of thoase ignition issues forever! I bought a new distributor cap, about 10 feet of HP-440 copper ignition wire, four screw-on spark suppressor resistors, and a bunch of inner-wall sealing heat shrink tubing. I took the distributor cap into the shop at work and milled a groove in the bottom edge where the cap seated against the distributor body, and made an O-ring gasket using (then brand-new) industrial-grade RTV silicone rubber. Then I assembled a new harness using the copper wire, suppressors, and shrink tubing. (The Rover already had full shroud gasketed spark plug covers, so that part of the system was already waterproof). I installed the new harness and cap, and drove that beast trouble-free (at least from the ignition standpoint) for 7 more years! Only time it ever stalled was one time when I got caught in a flash flood from a tornado. The engine ran fine until the water reached the air intakes!
Interesting problem to troubleshoot -- and one that reminds me of one encountered in the old Ford ignition systems of the 1960's. Back then those cars that were shipped with in-dash tachometers actually routed battery power through the tach to the ballast resistor that fed the primary side of the ignition coil. The whole setup was a series circuit, where one of the elements was the tach. Inside the tach was a small transformer with a handful of windings on one side and many thousands on the other, so that every time a pulse went through it as points opened and closed, there'd be a higher voltage, lower current pulse on the transformer windings on the other side. From the ignition coil's point of view, this small theft of energy didn't impact sparking in the least. But it did provide just enough energy that when diode clamped and integrated with a capacitor could drive a d'Arsonval movemen that moved the tach needle on the car's instrument panel.
The problem was that if any of the solder joints were open on the primary side of that transformer in the tachometer, the entire car died, and it became a real pill to chase back the source of the problem. Ford used plenty of solder, but nobody expected the cars to last long enough for grain-boundary migration of SnPb alloy to open up cracks and Kirkendall voids in the joint. And nobody anticipated that body styles like the Mustang, Cougar, and Thunderbird would become such classics that the cars would still be out there running around the streets three or four decades later -- which when you calculate e-KT for grain boundary migration times is about what you get.
In the early 1990's there was a rash of failures of these systems, and I always told people diagnosing their problems to look for the words "Sun-X" in the corners of the automotive glass as a tipoff that this could be a bad tachometer. When asked why, I'd explain that this was the indicator of "tinted glass," which was an option (not standard offering) in the 1960's that'd tip you off that the car was sold in a southern state -- where summertime interior temperatures in the passenger compartment drove up the "T" portion of that e-KT equation. After a few years, failures became less common, and then they became more common again -- only this time among cars without tinted glass. These were the "northern" cars where interior temperatures drove the PbSn grain boundary migration reaction more slowly.
Nowadays the occasional tach failure does show up, but most cars that have ever had it have had it fixed by now -- and with the mystery solved, car collectors either know what's happening or know somebody who can help troubleshoot. The problem is always, however, that a solder joint opening up is an intermittant open, which can always be a pill do to diagnose under any circumstances.
Smart collectors of these cars often just open the tachometer and reheat those solder joints as a matter of course the minute they buy one of these cars, anyway.
I must agree with Jacob ('73 AB3A DE KC2FHZ) while there were some microprocessors like the 4004 > 8008 > 8080 > Z80 (anyone remember dual clocks?) they were seldom used in mobile applications. Keep in mind that back then, computers were mostly analog. Analog circuitry would receive data as potential differences and pulse timings, and output "computed" data as voltages, frequencies and pulsewidths. Think summing (/difference) junctions, analog multipliers and log amps, variable reference comparators and PWMs. I (and a quarter of a million other people) owned early '70s Volkswagen Type IIIs and IVs (Fastback/Squareback) which included a very sophisticated Bosch analog high pressure fuel injection engine management system. The stock 1600cc Type I (Beetle) with a Solex carbureator developed 82 HP. The type IIIs with the same engine block developed 128 HP with about the same emissions and fuel consumption.
Vehicles of that era certainly did have spark control computers! Chrysler offered the electronic spark advance, and was working very hard on electronic fuel metering. The silicone and sand potting mixture was fairly effective, and the easy way to remove it was to soak the assembly in gasoline for a week or two and then carefully peel off the silicone. Note that it did not work with the uerethane based system that they changed to as a cost reduction. I had an old Dodge van with the 318 V8, and after 27 years it also developed an ignition failure mode. The source of the problem was different,however. IT had a replacement ignition module, one of those with the lone transistor on the gold heatsink. But on the replacement module that transistor was a dummy, theswitching was done by a power-tab TO-220 transistor inside the case. Unfortunately the poor encapsulating job had allowed water inside, and the steel case had rusted below the transistor leads. The rust crystals had formed a varistor that was effectively snubbing the ignition pulse so that there was not enough sperk energy to jump both the gap in the distributor and the gap in the plugs. By the time I discovered that, I had already replaced the module with another of the original style ones, which had the engine running again. But I saved the aftermarket module as an example of poor quality.
How, pray tell, could the described module with only power, ground, and an OUTPUT possibly do ANYTHING, much less adjust timing???? Even with input from the points (a 4th wire not described, or the "coil wire" served as I/O in parallel with the points), I don't see how the "computer" is getting any useful input for determining optimum timing based on engine load. I can imagine the best it could do was RETARD timing based on RPM (assuming the "parallel points" scenario). P.S. I had a 1972 Plymouth with the 318 V8 (see earlier blog post about its electrical problems).
This calls to mind assembly techniques used in a number of modern, miniaturized gadgets, where if you take them apart, you sometimes see that the rechargeable power cell isn't solidly soldered in place. Usually there's that thin, slowly corroding tab, attached with poorly fluxes solder. But more times than you'd think, the battery is just press-fit. Hey, don't drop that thing.
The story reminds me of my first computer, a Commordore PET. After about a month of operation it began to loose the sync to the display. I used a pencil eraser to gently push on the ICs on the circuit board and one popped out of the board. Commodore probably could not get the IC on time and built the board with plugs in the IC lead holes. When they finally got the part, they completely forgot about soldering it into the board. At least my problem was easily repaired without having to remove potting compound.
Timing was adjustable by just turning the whole ignition distributor, while looking a stroboscopic light shining on the crankshaft pulley, which had angle marks ("BTDC", ATDC": before and after TDC (Top Dead Center) and a fixed pointer...
The described module most probably wasn't an ignition computer, but just a Transistor module that served the purpose of limiting the usual breaker points degradation by the multiple sparking, which was supposed to be kept under control by the old type paper capacitor, usually found inside the distributor. Heat and Ozone from the heavy HV sparking inside the distributor made those capacitors change their value or fail completely, which caused the contact points to become pitted in a short time and causing an ignition failure; and the need to perform a "Tune-Up"... by that time, the carburetor must have been partly fouled by dirt and gum formation from the not-so-good quality (cleanliness) of the gasoline of that era. Therefore, the usual "Tune-Up" included a partial or complete disassemby and solvent washing of the carburetor on the fuel side, together with breaker points change and "calibration", "condenser" (capacitor) change, timing with the "timing light", sparkplug change (or at least cleaning with an old "Champion" sand blasting machine), regapping them; and a typical air and fuel filters change for complete service. This was required every 6 months or so, depending on the quality of the replacement parts, and frequently, the knowledge or the Tune-Up mechanic... Good ones carefully re-gapped every spark plug to strict values, keeping a close look with a powerful loupe, patiently adjusting the ground electrode to be correctly bent, and took the extra time and effort to lubricate all the bushings and slide points in the breaker points plate, checking the vacuum diaphagm and finally placing a little bit of high temperature grease on the distributor cam, so that the new breaker points kept their opening gap much longer!
I have a 1967 Ford Falcon left to me by my father, still running fast, but previously had to pay a lot for the replacement breaker points for the much vaunted "Mallory" Dual point centrifugal advance distributor... those lasted even less than 6 months of daily driving, until someone told me about the small transistor box, which was made in Mexico by an old, very knowledgeable Doctor in Physics as a side business, NO MORE need to replace the expensive Mallory dual point breaker points anymore! By lubricating the distributor cam, the last points set I bought has lasted more than 15 years! (That box used a rugged Toshiba TV horizontal transistor, rated for high voltage, and has endured all these years without a hitch, reducing the current jumping through the breaker points from several amperes to less than 4 miliamperes, which keeps the surface of the contact points clean and as good as new!).
By the way, my father had both an International Scout-800 (a kind of 'jeep') AND a Ford Bronco, both of 1966 vintage... the Scout by International proved to be a much better, rugged, durable and trouble free vehicle, even at a lower sales price! I Actually learned to drive on it. The closed cabin was much more comfortable than the canvas cover of a Jeep, and the overall quality of the Scout was a notch above the others. The only inferior aspect were the front wheel "locks" which required the driver to come down the vehicle to be able to engage the 4-wheel drive in order to turn both front wheels "locks" manually. Fond memories!
Ohh, I forgot to tell that about after five years fo using that transistor box with my Dual-Points distributor, it was the box that failed during a very wet weekend driving at high speed in highway... The culprit was the epoxy potting becoming cracked and allowing water to enter and corroding the TO-3 Transistor socket on the two terminal pins. As I became friend with the Physicist that fabricated it, he was kind enough to replace it at no charge, but it was interesting seeing HOW he solved the failure: He traced the failure to the Epoxy potting compound shrinking over time, and developing deep cracks. The thermal expansion of the components cycling, plus the epoxy clinging to the walls of the aluminum housing facilitated this.
The solution was elegantly simple: heat the aluminum housing to a temperature slightly above the upper limit of the operating range, change the epoxy to a different type that cured with heat, and let the potted housing cool down as the Epoxy hardened. Thus, the potting compound and assembly were left in a slight COMPRESSION state, which eliminated the tendency of cracking altogether. The second unit has been working for more than 10 years and I can't see the slightest trace of cracking!
there are actually four ignition timing advance schemes that have been implemented -sometimes in combination- over the years. (1) A manual lever on the steering columns (any octogenerians out there?), (2) vacuum advance, (3) mechanical (centrifugal) advance and (4) electronic delay line.
Here's the most amazing part of this story: "The connection was barely touching the edge of the pad." Without any evidence of solder, the fact that it held together for 27 years is inconceivable.
The description of the failing ignition mystery was exactly how my 1961 Ford Fairlane acted. I was given the Fairlane in the fall of 1971 to use as my wintertime 'beater' so as to keep my 1971 Hemicuda muscle car stored during the Northeast Ohio snow bound season. And as Lee described, the Fairlane's engine would simply quit running at random. Quite often I could shift into neutral and turn the key to restart the engine while still rolling - pop back into 'drive' and keep going. The ignition failure might occur within the next 1/2 mile or not for days. Much troubleshooting and ignition parts replacement did not solve the problem. About a year later I gave the Fairlane to my Brother-in-law - he had hopes of solving the problem himself but ended up having the car towed to the junk yard. Years later, I described the problem to an Uncle, and he knew immediatley what the problem was and how to fix it ! Down on the steering column, under the dash board was a switch mounted with a movable contact attached to the shift rod. The purpose of the switch was to prevent cranking the starter motor when the shift lever was in any position except PARK or NEUTRAL. The switch was about 2 - 3 inches long and curved to conform around the surface of the steering column. The switch was mounted with two screws, one on each end of the switch housing. The holes were elongated to allow adjustment of the switch position to match the shift lever indicator. Wouldn't you know that the screws came loose, and the switch would rotate out of position while driving, cutting the ingnition off. By moving the shift lever to NEUTRAL, (or PARK if stopped) the switch would be pushed back into correct alignment and away we'd go ! - until the next time...
Some years ago I drove a 1968 Pontiac 4-dr hardtop with the regular fuel 400 cid engine. I drove these cars for more than 30 years.
I bought my first one in 1969 near Pontiac, Michigan. All of these cars were extremely reliable. Maybe the second or third car I drove started to stall out with about 70K or so on the odometer. I would be driving along and the engine would just stop.
I would get out, open the hood and move the wires around, looking for a problem. When I got back in the car to try to start the engine, it would start and run fine until the next time.
This happened for a period of maybe 3 months. Finally, at home, I decided to take a closer look. The primary ignition wire had been draped over a clip on the firewall and over time the wire insulation was worn away so that under just the right conditions the wire would ground itself and the engine would stop. When I reached under the hood and wiggled the wires it was just enough to move the wire away from the clip and remove the short.
The worn insulation was hidden behind some tubes that were also supported by the clip. A piece of electrical tape fixed the problem for good. I wish today's cars were as simple.
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 5
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This radio show will show what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
7 
